Jean Wu

Jean Wu is a natural language processing researcher and Stanford Symbolic Systems alum who coauthored the paper behind the Stanford Sentiment Treebank. The 2013 paper "Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank" had seven authors, introduced a treebank of 215,154 labeled phrases and the Recursive Neural Tensor Network, and came with a live demo that Wu co-created with Richard Socher, Rukmani Ravisundaram and Tayyab Tariq. In 2014, they coauthored an EMNLP paper on relation extraction with Gabor Angeli, Julie Tibshirani and Christopher Manning. The sentiment paper won the ACL Test-of-Time Paper Award for 2013 papers, announced at ACL 2023. As of October 2026, they work at Apple.

Now at: Apple

Timeline

2013

The Sentiment Treebank paper

"Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank", published at EMNLP 2013, has seven authors, among them Jean Wu. The paper introduces the Stanford Sentiment Treebank, with fine-grained labels for 215,154 phrases in 11,855 sentences, and a Recursive Neural Tensor Network that lifts single-sentence positive/negative accuracy from 80% to 85.4%.
2013

A live sentiment demo

In 2013, the project website runs a live demo that predicts the sentiment of movie reviews. Richard Socher, Rukmani Ravisundaram and Tayyab Tariq join Wu in building it.
2014

Relation extraction at EMNLP

At EMNLP 2014, Gabor Angeli, Julie Tibshirani, Wu and Christopher Manning publish "Combining Distant and Partial Supervision for Relation Extraction".
2023

An ACL Test-of-Time award

The 2013 sentiment treebank paper wins the ACL Test-of-Time Paper Award for 2013 papers, announced at ACL 2023. The winners are its seven authors: Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher Manning, Andrew Ng and Christopher Potts.
As of October 2026

Working at Apple

Among the Stanford NLP Group's Symbolic Systems alumni, Jean Wu works at Apple.